The vertical limit for randomised trials

Alison Campbell considers the evidence for the efficacy of parachutes.

Recently a teacher sent me a paper titled: ‘Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials‘ (Smith and Pell, 2003, BMJ 327: 1459-1460). I have to say I chuckled when I read this – a common charge levelled against current medical practice by the alternative health lobby is that many medical techniques haven’t been subjected to randomised controlled trials (with the corollary that it’s thus unfair to demand evidence from such trials on alternative practices).

The authors state they conducted a literature search of some of the major science sources, using the search words ‘parachute’ and ‘trial’. However (and unsurprisingly), they found no randomised controlled trials (RCTs) of parachute use. Smith and Pell begin their discussion with the following inspired statement:

“It is a truth universally acknowledged that a medical intervention justified by observational data must be in want of verification through a randomised controlled trial.”

Many medical interventions probably fall into this category – for example, I doubt that surgery for severe appendicitis has ever been subjected to such a trial. That’s not to say that, where appropriate (and in the case of appendicitis it almost certainly isn’t!) such trials shouldn’t be performed. As Smith and Pell point out, hormone therapy for post-menopausal women seemed – on the basis of observational studies – to convey a number of health benefits. But RCTs showed that hormone replacement therapy actually increased the risk of ischaemic heart disease.

As the authors say, RCTs avoid a major weakness of observational studies: that of bias (eg selection bias and reporting bias). They note that individuals jumping from aircraft without the help of a parachute are likely to have a high prevalence of pre-existing psychiatric morbidity (ie they are probably not in their right minds when they jump. You have got to love this paper!). So any study of parachute use could well be subject to selection bias, in that those using them are likely to have fewer psychiatric problems than those who don’t. Smith and Pell also put forward the possibility that enforced parachute use is simply a case of mass medicalisation of the population by out-of-control doctors – or worse, by evil multinational corporations. (These are, of course, charges frequently levelled at the medical world, eg by those who are against interventions such as vaccination.)

This little gem of a paper contains some valuable lessons on the nature of science (and more particularly, science-based medicine). And it should be read by anyone who doubts that scientists have both creativity and a good sense of humour.

Keeping it in perspective


The promotion of critical thinking can seem an uphill struggle, but at least we don’t get torn limb from limb for trying.

As skeptics, we fight to uphold certain freedoms: freedom of enquiry, freedom of speech, academic freedom. The battles are worth waging to enjoy the blessings these principles bring, including knowledge and understanding, and because the alternative is darkness.

Skeptics skirmish in public debates, across the kitchen table, and in the workplace. While I think every academic should be a skeptic, I work alongside lecturers who oppose our organisation and who are hostile to science. Some of them believe in Creationism, homeopathy, or the Book of Mormon as a divinely inspired history of ancient America. Such lecturers exasperate me. University managers who do not appreciate academic freedom irk me. Being a skeptic can be tiring and frustrating.

Still, we need to keep our battles in perspective. In parts of the world, debating a university speech code is the farthest thing from the concerns of millions of people. For example, the girls and women of Afghanistan would be overjoyed to have any education at all. Five years after being ousted by the United States, the Taliban still control large parts of Afghanistan. They invoke societal codes in the name of custom and religion as justification for denying women their rights, including the right to an education.

The director of education at Ghazni, Fatima Mustaq, says she has received death threats for refusing to send girls home from school. The threats are also against her husband and their eight children. During the Taliban’s rule, she and her sister secretly taught girls at their home. “They found out and raided us. We managed to persuade them that we were only teaching the Koran. But they spied and found out we were teaching algebra. So they came and beat us. Can you imagine, beating someone for teaching algebra?”

In November, gunmen came for Mohammed Halim and dragged him from his home at Ghazni while his children cried and his wife begged for mercy. The 46-year-old schoolteacher was then partly disembowelled before being torn apart with his legs and arms tied to motorbikes. His remains were put on display as a warning to others to stop educating girls. Halim was one of four teachers killed in rapid succession at Ghazni for defying a Taliban order to not teach girls.

Before they can hope to gain an education, Afghan women and girls need protection from abduction and rape by armed men, being traded to settle disputes and debts, and forced and underage marriage. By being married as young as 12, females are denied their right to education and the freedom to decide the course that they wish their lives to take. A ban on interaction between unrelated men and women greatly inhibits women’s access to the workplace, courts, and schools, because these places are segregated or exclusively male.

Afghan women get almost no protection from the state. In fact, a report issued in November by the Pentagon and the US State Department found that the police force in Afghanistan is largely incapable of carrying out routine law enforcement. Violence against Afghan women is normal. It is tolerated at the highest levels of government. Some judges disregard the law and rely on tradition to hold women responsible for being attacked, thus sanctioning brutality. The perpetrators of violence against women are rarely charged. If cases are prosecuted, the men are usually let go or punished lightly. Women who report rape risk being accused of having committed the crime of having sex outside marriage.

Violence against women by family members also is common. It ranges from forced deprivation of education to beatings, sexual violence and killings. Many acts of violence involve traditional crimes of honour, when a female is punished by her family for shaming them; perhaps she got raped. Punishment can mean being stoned or burned or beaten to death.

Glimmers of progress are visible in Afghanistan. In 2004 a new constitution was adopted which proclaimed that “the citizens of Afghanistan-whether man or woman-have equal rights before the law.” It also provides for a minimum representation of women in both houses of parliament.

In 2005 the first woman was appointed as governor of a province. Over 40 percent of women were registered as voters in 2004. Women are officially allowed to seek employment-albeit with permission from family members. One in five girls now attends primary school. Nearly all the younger women interviewed recently by Amnesty International expressed their wish for the future as simply being able to continue their education. As skeptics in New Zealand look to soldier on in 2007, we can be thankful for the freedoms we enjoy.